Human genetic studies have identified a small number of high confidence causal genes for autism spectrum disorder (ASD). Even for these highly penetrant mutations, there are individual differences in ASD-relevant symptoms and other co-occurring behavioral and medical phenotypes [e.g. anxiety, cognitive deficits, gastrointestinal (GI) disturbances]. Symptom heterogeneity diminishes the ability to predict response to treatments, but presently there is limited understanding of how genomic risk variants, genetic background and environmental factors impact ASD risk and clinical heterogeneity. Animal genetic reference panels provide opportunities to systemically study genetic risk variants in relation to a panel of diverse, deep sequenced and reproducible recombinant inbred (RI) genetic backgrounds. This strategy is highly relevant to discovery of human disease etiology, currently applied mostly to experimental models of non-brain disorders, including susceptibility (and reduced vulnerability) to infectious diseases, toxins, cancer, and respiratory, GI and metabolic disorders. Here, we propose to use this strategy to address the knowledge gaps in understanding ASD susceptibility and phenotype heterogeneity. As background, chromodomain helicase DNA binding protein 8 (CHD8) encodes a chromatin remodeling protein that is a very high confidence ASD-risk gene, accounting for ~0.5% of cases. Macrocephaly and ASD are expressed in nearly all subjects with CHD8 mutations, but symptoms can vary in levels of impairment and even expression. In mice, homozygous null mutations in Chd8 are embryonic lethal. Heterozygous null mice (Chd8+/-) recapitulate macrocephaly, but exhibit varying degrees of emotional dysregulation, repetitive behavior, social communication and learning deficits, all of which has been done on a single background (C57BL/6) strain. The proposed experiments leverage the phenotypic features associated with CHD8 haploinsufficiency in humans and mice with our laboratory?s recent studies showing that RI mouse strains express broad heterogeneity and high heritability of social and cognitive traits. Proposed experiments will examine how strain genetic diversity imposes susceptibility (or reduced vulnerability) to the expression of social, cognitive and emotional phenotypes, as well as macrocephaly and GI disturbances, in Chd8+/- mice. Using the Collaborative Cross (CC), a highly genetically diverse and fully sequenced RI mouse panel, the studies provide an exciting opportunity to reveal the effects of genetic diversity on ASD-relevant phenotypes. Aim 1 will determine the impact of genetic background on behavioral and physiological phenotypes in a newly generated CC/WT and CC/Chd8+/- mouse panel. Aim 2 will apply RNA- sequencing in frontal cortex and dorsal hippocampus to identify gene networks that underlie social and cognitive deficits in the most susceptible CC/Chd8+/- strains. This work advances the power of preclinical experimental studies, whole-genome sequencing, and deep behavioral and molecular phenotyping, to better understand the heterogeneity of ASD.